Testing kernel density reconstruction for Lagrangian photochemical modelling

In recent years, a number of pioneering works have shown as Lagrangian models can be of great interest when dealing with photochemistry, provided that special care is given in the reconstruction of chemicals concentration in the atmosphere. Density reconstruction can be performed through the so-called “box counting” method: an Eulerian grid for chemistry is introduced and density is computed counting particles in each box. In this way one of the main advantages of the Lagrangian approach, the grid independence, is lost. In this paper, a new approach to Lagrangian photochemical modelling is investigated and the chemical module of the Photochemical Lagrangian Particle Model (PLPM) is described and fully tested for stability, reliability and computational weight. Photochemical reactions are treated in PLPM by means of the complex chemical mechanism SAPRC90 and four density reconstruction methods have been developed, based on the kernel density estimator approach, in order to obtain grid-free accurate concentrations.